Peer-to-Peer lending platforms may lead to cost reduction, and to an improved user experience. These improvements may come at the price of inaccurate credit risk measurements, which can hamper lenders and endanger the stability of a financial system. In the article, we propose how to improve credit risk accuracy of peer to peer platforms and, specifically, of those who lend to small and medium enterprises. To achieve this goal, we propose to augment traditional credit scoring methods with “alternative data” that consist of centrality measures derived from similarity networks among borrowers, deduced from their financial ratios. Our empirical findings suggest that the proposed approach improves predictive accuracy as well as model explainability.
Network based credit risk models
Giudici P.;Spelta A.
2020-01-01
Abstract
Peer-to-Peer lending platforms may lead to cost reduction, and to an improved user experience. These improvements may come at the price of inaccurate credit risk measurements, which can hamper lenders and endanger the stability of a financial system. In the article, we propose how to improve credit risk accuracy of peer to peer platforms and, specifically, of those who lend to small and medium enterprises. To achieve this goal, we propose to augment traditional credit scoring methods with “alternative data” that consist of centrality measures derived from similarity networks among borrowers, deduced from their financial ratios. Our empirical findings suggest that the proposed approach improves predictive accuracy as well as model explainability.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
